Distributed Leadership in Organizations a Review of Theory

International Journal of Management Reviews, Vol. 13, 251–269 (2011) DOI: 10.1111/j.1468-2370.2011.00306.x Distributed Leadership in Organizations:

A Review of Theory and Researchijmr_306 251..269

Richard Bolden Centre for Leadership Studies, University of Exeter Business School, Rennes Drive, Exeter EX4 4ST, UK Corresponding author email: [email protected]

The aim of this paper is to review conceptual and empirical literature on the concept of distributed leadership (DL) in order to identify its origins, key arguments and areas for further work. Consideration is given to the similarities and differences between DL and related concepts, including ‘shared’, ‘collective’, ‘collaborative’, ‘emergent’, ‘co-’ and ‘democratic’ leadership. Findings indicate that, while there are some common theoretical bases, the relative usage of these concepts varies over time, between countries and between sectors. In particular, DL is a notion that has seen a rapid growth in interest since the year 2000, but research remains largely restricted to the field of school education and of proportionally more interest to UK than US-based academics. Several scholars are increasingly going to great lengths to indicate that, in order to be ‘distributed’, leadership need not necessarily be widely ‘shared’ or ‘democratic’ and, in order to be effective, there is a need to balance different ‘hybrid configurations’ of practice. The paper highlights a number of areas for further attention, including three factors relating to the context of much work on DL (power and influence; organizational boundaries and context; and ethics and diversity), and three methodological and developmental challenges (ontology; research methods; and leadership development, reward and recognition). It is concluded that descriptive and normative perspectives which dominate the literature should be supplemented by more critical accounts which recognize the rhetorical and discursive significance of DL in (re)constructing leader– follower identities, mobilizing collective engagement and challenging or reinforcing traditional forms of organization.

Introduction Distributed leadership has become a popular ‘post-heroic’ (Badaracco 2001) representation of In an article entitled ‘Distributed properties: a new leadership which has encouraged a shift in focus architecture for leadership’, Peter Gronn (2000) out- from the attributes and behaviours of individual lined the concept of ‘distributed leadership’ (DL) as ‘leaders’ (as promoted within traditional trait, situ- a potential solution to the tendency of leadership ational, style and transformational theories of thinking to be divided into two opposing camps: leadership – see Northouse 2007 for a review) those that consider it largely the consequence of indi- to a more systemic perspective, whereby ‘leader- vidual agency (e.g. Bass 1985) and those that present ship’ is conceived of as a collective social process it as the result of systems design and role structures emerging through the interactions of multiple actors (e.g. Jaques 1989). Over the subsequent decade, as (Uhl-Bien 2006). From this perspective, it is the papers in this Special Issue testify, the concept of argued: DL has gone from strength to strength and has made substantial inroads into particular areas of theory and Distributed leadership is not something ‘done’ by practice. an individual ‘to’ others, or a set of individual

© 2011 The Author International Journal of Management Reviews © 2011 British Academy of Management and Blackwell Publishing Ltd. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA 252 R. Bolden

actions through which people contribute to a group namebutafew.1 Common across all these accounts or organization...[it] is a group activity that works is the idea that leadership is not the monopoly or through and within relationships, rather than indi- responsibility of just one person, with each suggest- vidual action. (Bennett et al. 2003, p. 3) ing a similar need for a more collective and systemic understanding of leadership as a social process For Gronn (2002), DL offered the promise of (Barker 2001; Hosking 1988). a new ‘unit of analysis’ through which leadership Within this paper I explore the lineage of the could be understood in a holistic sense rather than concept of DL and its recent rise to prominence. I simply as the aggregation of individual contribu- explore the emergence of DL as a discrete body of tions. He referred to this dimension of leadership as literature, reflecting on its relative use in relation to ‘concertive action’ (as contrasted with ‘numerical alternative conceptions of leadership as a shared action’) and illustrated his argument with three process. I review the main theoretical developments alternative forms of engagement (‘spontaneous col- in this field and the manner in which these ideas have laboration’, ‘intuitive working relationships’ and been embraced and applied within different sectors ‘institutionalized practices’) each of which could be and contexts. I explore the empirical evidence from considered as a manifestation of ‘conjoint agency’. research into DL within organizations, highlighting In setting out his argument, Gronn called for a fun- some common themes and areas of difference. damental reframing of leadership, suggesting that it Finally, I reflect on the direction in which the field ‘is more appropriately understood as a fluid and seems to be headed and priorities for further investi- emergent, rather than as a fixed, phenomenon’ gation.2 (Gronn 2000, p. 324) – a call that has been enthu- siastically received by scholars and practitioners alike. The theoretical origins of DL In taking a distributed perspective, attention turns from generic accounts of the attributes and/or actions While it is only really since the turn of the millen- of individual leaders to ‘situated leadership practice’ nium that the concept of DL has been widely (Spillane 2006). According to Spillane and Diamond embraced by scholars and practitioners, the origins (2007b, p. 7) ‘a distributed perspective on leadership of the concept go back quite a bit further. Oduro involves two aspects – the leader plus aspect and the (2004, p. 4) suggests that accounts of DL date back practice aspect’. The ‘leader-plus’ aspect ‘acknow- as far as 1250 BC, making it ‘one of the most ancient ledges and takes account of the work of all the leadership notions recommended for fulfilling organ- individuals who have a hand in leadership and man- izational goals through people’. In terms of its theor- agement practice’ rather than just those in formally ization, however, Harris (2009, p. 3) proposes that it designated ‘leadership’ roles. The ‘practice’ aspect ‘is an idea that can be traced back as far as the mid ‘foregrounds the practice of leading and managing 20s and possibly earlier’. Gronn (2000) cites Gibb [...and...] frames it as a product of the interac- (1954) as the first author to refer explicitly to DL tions of school leaders, followers, and aspects of when proposing that ‘leadership is probably best their situation’. According to these authors, together conceived as a group quality, as a set of functions these aspects of leadership offer an analytical frame- which must be carried out by the group’ (Gibb 1954, work for ‘examining the day-to-day practice of lead- cited in Gronn 2000, p. 324). Gibb’s distinction ership and management’ rather than dwelling on ‘leaders and leadership structures, functions and 1Additional terms include ‘dispersed’ leadership’ (e.g. roles’. Gordon 2002, 2010; Ray et al. 2004) and ‘distributive lead- Distributed leadership, however, is not the only ership’ (e.g. Brown and Gioia 2002), although in each case these are used far less widely than those terms selected for theory or approach to call for such a reframing of investigation in this paper. how we understand leadership. The notion of ‘shared 2The literature for this paper was identified from a compre- leadership’ (SL) has also been in use for some time hensive search of bibliographic databases (including Busi- (see Pearce and Conger 2003a for a review), as ness Source Complete, EBSCO, JSTOR and Scopus) as well have those of ‘collective leadership’ (e.g. Denis et al. as analysis of a number of recent reviews (including Harris 2009; Leithwood et al. 2009a; and Spillane and Diamond 2001), ‘collaborative leadership’ (e.g. Rosenthal 2007a). As indicated in the title, the primary focus of the 1998), ‘co-leadership’ (e.g. Heenan and Bennis paper is on DL in organizations rather than, for example, in 1999) and ‘emergent leadership’ (e.g. Beck 1981), to communities or other groups.

© 2011 The Author International Journal of Management Reviews © 2011 British Academy of Management and Blackwell Publishing Ltd. Distributed Leadership in Organizations 253 between ‘two forms of distribution: the overall such as: Mary Parker Follett’s (1942/2003) work on numerical frequency of the acts contributed by each reciprocal influence; Benne and Sheats (1948) work group member and “the multiplicity or pattern of on the diffusion of leadership functions within group functions performed” ’ (Gronn 2000, p. 324) groups; Gibb’s (1954) work on leadership; French form the basis for Gronn’s distinction between and Snyder’s (1959) and Dahl’s (1961) work on the numerical and concertive action and provide the fun- distribution of power and influence; Becker and damental building blocks for subsequent theoretical Useem’s (1942) and Etzioni’s (1965) work on dual development. leadership; Kerr and Jermier’s (1978) work on sub- Despite this early interest in the concept, however, stitutes for leadership; Katz and Kahn’s (1966, 1978) as Gronn (2000, p. 324) suggests, the idea of DL ‘lay work on sharing leadership; and Schein (1988) on the dormant until its resurrection by Brown and Hosking functions of leadership. (1986)’ and is only mentioned in a smattering of In addition to these sources Harris (2009) cites the articles during the 1980s and 1990s (e.g. Barry 1991; work of: Festinger et al. (1950) and Heinicke and Beck and Peters 1981; Gregory 1996; Leithwood Bales’ (1953) on informal leadership in groups and et al. 1997; Senge 1993) – most probably owing to teams; Barnard (1968) on the functions of the execu- the appetite for accounts of ‘new leadership’, tive and the informal organization; Manz and Sims’ founded on ‘transformational’ and/or ‘charismatic’ (1993) social learning theory; Hutchins’ (1995) work leadership by senior executives, that dominated on distributed cognition and ‘lateral agency’; Louis scholarly and practitioner literature during this and Marks’ (1998) work on professional learning period (see Parry and Bryman 2006 for further dis- communities; and Wheatley’s (1994) work on com- cussion). Prior to and during this period, however, plexity and systems. while specific references to DL may have been few Finally, Leithwood et al. (2009b) highlight the sig- and far between, some significant conceptual devel- nificance of organizational learning theory (Hutchins opments were achieved which, in many ways, paved 1995; Weick and Roberts 1993); distributed cogni- the way for later work. tion (Jermier and Kerr 1997; Perkins 1993; Salomon In tracking the theoretical origins of DL, a number 1993); complexity science (see Uhl-Bien et al. 2007, of key concepts are commonly cited. Spillane et al. and Osborn and Hunt 2007 for a review); and ‘high (2004), for example, identify ‘distributed cognition’ involvement leadership’ (see Yukl 2002). and ‘activity theory’ as the conceptual foundations of Together, these authors map out a rich and diverse their particular account of DL. The first of these array of theory and research upon which subsequent concepts represents human cognition and experience work on DL can build, which resonates closely with as integrally bound up with the physical, social and the theoretical origins of SL theory as outlined by cultural context in which it occurs (see, for example, Pearce and Conger (2003b). Despite this lineage, Hutchins 1995; Latour 1987; Lave and Wenger 1991; however, it was not until the mid-1990s that ‘condi- Leont’ev 1981; Pea 1993; Resnick 1991). The second tions were finally right for the acceptance of this approach highlights the manner in which human seemingly radical departure from the traditional view activity is both enabled and constrained by indi- of leadership as something imparted to followers by vidual, material, cultural and social factors (see, for a leader from above’ (Pearce and Conger 2003b, example, Brown and Duguid 1991; Giddens 1979, p. 13). Pearce and Conger offer a number of reasons 1982; Vygotsky 1978; Wertsch 1991). for this shift, including the rise in cross-functional Gronn (2000) similarly recognizes distributed teams, along with speed of delivery, the availability cognition and activity theory as key concepts within of information, and greater job complexity. Lipman- DL. With regard to activity theory, he draws particu- Blumen (1996) also cites increasing global inter- larly on the work of Engeström (1999) (who, in turn, dependence and demands for inclusion and diversity builds on the work of authors such as Vygotsky 1978, as driving factors that highlight the limitations of and Leont’ev 1978, 1981) which offers a framework more individualistic understandings of leadership. In for analysing situated activity as the product of recip- effect, it is argued, the leader-centric approach which rocal and mediated interactions between ‘instru- worked well enough and offered a (perhaps illusory) ments’, ‘subjects’, ‘objects’, ‘rules’, ‘community’ promise of order and control that suited organiza- and ‘division of labour’. tions (or their directors and shareholders at least) Gronn (2008a), however, also cites the signifi- throughout much of the 20th century is no longer fit cance of a stream of additional theory and research, for purpose and needs to be revised.

Despite this, the idea of the individual leader spurt that would do any teenager proud’ (Leithwood remains hard to shed and, to a large extent, we remain et al. 2009c, p. 269). enamoured of the ‘romance of leadership’ (Meindl A search of google.co.uk on 8 March 2011 1995; Meindl et al. 1985) whereby organizational returned 187,000 hits for the phrase ‘distributed actors and observers tend to over-attribute perfor- leadership’ and books.google.co.uk revealed 9,220 mance outcomes to the contribution of individual books referring to the topic. While this is only a ‘leaders’, while disregarding other equally, if not small fraction of the overall literature on ‘leadership’ more, important factors. As O’Toole et al. (2003, (201 million web pages and nearly six million books p. 251) suggest: ‘shared leadership for most people is on google.co.uk) if considered alongside related lit- simply counterintuitive: leadership is obviously and eratures, such as ‘shared’, ‘collective’, ‘collabora- manifestly an individual trait and activity’. They tive’, ‘co’ and ‘emergent’ leadership, it represents a illustrate this paradox through reference to leaders significant and growing body of material. such as Gandhi and Luther King, Jr, proposing that: Website statistics also suggest, however, that DL is ‘when the facts are fully assembled even the most less prevalent as a concept within common usage fabled “solitary” leaders relied on the support of a than shared, collaborative or collective leadership. To team of other effective leaders’. Such a deeply this extent, it is interesting to explore the domains in entrenched tendency to underestimate the contribu- which DL has been embraced as a way of describing tion of more than a few key figures, it is argued, shared/dispersed forms of leadership and where it ‘stems from thousands of years of cultural condition- has made less headway. ing’ and, as such, remains incredibly difficult to In order to analyse the comparative growth of the change, even if the evidence points elsewhere. field of DL, it was decided to focus on academic Indeed, as Astley (1985, p. 503, cited in Alvesson articles as an indicator of scholarly activity. Using 1996, p. 475) suggests, ‘theories gain favour because information from the Scopus database (one of the of their conceptual appeal, their logical structure, or largest abstract and citation databases of its kind) their psychological plausibility’ rather than their Figure 1 summarizes the number of publications on empirical accuracy per se. concepts related to DL (based on a search of title, Such an observation is helpful when accounting abstract and keywords for papers classified as for the initial slow uptake of distributed and shared ‘reviews’ or ‘articles’) between 1980 and 2009 perspectives on leadership, but also cautionary in (http://www.scopus.com, accessed 26 January 2010). encouraging reflection on why these concepts may be While I in no way wish to imply that this graph gaining quite so much attention nowadays. Do they includes everything published on these concepts offer a genuine alternative to earlier conceptualiza- during this period, it does give an illustration of the tions or are they simply ‘the emperor’s new clothes’? relative proportion of articles on each concept over Do they offer an accurate account of how leadership time. Figure 1 clearly shows that DL has seen a rapid actually occurs or simply respond to a current increase in profile since the year 2000, so much so demand within society for a greater sense of equity that for the last three years of the analysis period and purpose? (2007–09) it had overtaken SL as the term of preference for describing such forms of leadership. Interest in DL, however, is a recent phenomenon compared Distributed leadership: An idea whose with SL (which has seen a steady stream of publica- time has come? tions since the early 1990s) and ‘emergent leadership’ (which has seen a small but consistent trickle of Despite an initial resistance to the idea of DL, given articles since 1980). ‘Collective’ and ‘collaborative’ the changing nature of work and increasing disillu- leadership are concepts which have received ongoing sionment with the manner in which ‘new leadership’ interest since the mid-1990s (although not to the approaches (such as transformational and charis- same extent as either SL or DL), and ‘co-leadership’ matic leadership) glorify ‘heroic’ accounts about is a concept that mirrored these trends until the mid- senior executives, it would, indeed, seem that DL is 2000s and since then has declined. Thus, while it is ‘an idea whose time has come’ (Gronn 2000, p. 333) evident that overall interest in shared/dispersed – no longer ‘the new kid on the block’ (Gronn 2006, forms of leadership has seen a marked increase since p. 1) but rather ‘an area of study in an adolescent the year 2000, not all variants have achieved the same stage of development [. . .] experiencing a growth degree of attention.

Figure 1. Publications on Scopus database

To gain a clearer understanding of these dynamics, Table 1. Articles on shared and distributed leadership by national it was decided to conduct two further comparisons on affiliation of first author (Source: Scopus.com, for period these trends: (1) an analysis by national affiliation 1980–2009) of first authors, and (2) an analysis by subject Affiliation of Concept discipline.3 1st author From the first of these analyses, it became appar- Shared leadership Distributed leadership ent that, while the vast majority of articles are pub- US 103 articles 35 articles lished by US-based authors (230 out of 418 articles; UK 9 articles 47 articles as opposed to 70 from the UK, 34 from Canada, 14 Chi2 = 55.99, P Յ 0.01. from Australia, 10 from Denmark and all other countries with fewer than 10), there are some important differences in the relative popularity of terms by nursing and medicine; SL and emergent leadership country. In particular, the proportion of publications within psychology; and DL within business, man- on DL is significantly higher in the UK than in the agement and other areas of the social sciences (fol- US and vice versa for publications on SL (see lowed by SL). To elaborate further on these sector Table 1). This is an interesting trend and points to a differences, an analysis of source publications was US/UK divide between the two sets of literature. conducted from which it was identified that: The second analysis revealed some further inter- • 68% of DL articles were published in education/ esting trends in terms of the relative prevalence of educational management journals, compared with concepts between academic disciplines. In particular, only 22% of SL articles. Of these, 26% of the DL SL appears to be the concept of preference within articles were published in School Management and Leadership, with the remainder spread across a 3Owing to space constraints, graphs of these analyses are not range of sources. presented here, but can be obtained from the author on • 39% of SL articles were published in health- request. related journals (including nursing and medicine)

in comparison with no DL articles. Nine out of This, however, I feel would be unhelpful, owing the these 36 articles were published in the Danish manner in which authors either tend to use these nursing magazine Sygeplejersken. terms interchangeably, or go to great lengths to dis- • 25% of SL and 19% of DL articles were published tinguish between them. In either case, any attempt at in general business, management and leadership providing a definitive definition would fail to capture journals. Of these, the majority (eight articles each the complexity, and inherent paradoxes, of the field on DL and SL) were in the Leadership Quarterly. and would potentially foreclose a series of ongoing • The remaining articles were spread across a range debates and discussions that are both inevitable and of subjects, including public administration, desirable within an inchoate field of study such as general social science, engineering, computing this. Within this section, however, I consider a and psychology, with some differentiation between number of different concepts relating to the distribu- DL and SL terminology. tion of leadership within organizations, and the manner in which they tend to be used. Together, these findings indicate that, while the Leithwood et al. (2009b, p. 1) suggest that, for the concept of DL has made substantial headway in the majority of authors, DL can be considered to incor- past decade, its popularity remains very much porate shared, democratic, dispersed and other restricted to particular geographic and sector areas. related forms of leadership. From this perspective, While it may be relatively easy to speculate on the DL tends to be considered from a normative perspec- reasons for these differences (for example, the fact tive, as a means for enhancing the effectiveness of, that the term DL appears to have been picked-up and engagement with, leadership processes. For such and promoted within UK education policy and prac- authors, the key question is how leadership should be tice through the work of the National College for distributed in order to have the most beneficial effect School Leadership, thereby acquiring greater cur- (usually measured in terms of student learning out- rency within this context), a number of questions comes for research within schools). are raised which merit further attention. First, does A number of other authors (including Spillane and this differentiation suggest an unnecessary prolifer- Gronn), however, take an explicitly descriptive ation of terms (spearheaded by authors who are either approach, in which they argue that DL offers an unaware of, or who choose to ignore, comparable analytical framework through which one can assess work from elsewhere) or provide evidence of subtle and articulate the manner in which leadership is (and differences in the ways in which leadership is con- is not) distributed throughout organizations. Such ceived and enacted in different contexts? What authors go to great lengths to argue that, while lead- might these differences say about the discursive sig- ership may be shared and/or democratic in certain nificance of ways in which leadership is talked situations, this is not a necessary or sufficient about in particular contexts (for example, is there requirement for it to be considered ‘distributed’. Fur- something about schools and/or British people that thermore, they suggest that DL is not an alternative makes them inherently more responsive to the or replacement for individual/focused leadership and notion of ‘distributed’ leadership, and about health- that distribution per se is not necessarily related to care organizations and/or Americans that makes more effective or efficient leadership. As Spillane them engage with the idea of ‘shared’ leadership)? and Diamond (2007b, p. 11) put it, leaders ‘don’t And what might be done to consolidate this diver- have to see eye-to-eye or even have to get along with sity of terminology in order to offer a basis for one another to co-perform leadership routines and greater conceptual clarity and comparison between tasks...Whether two or more leaders seek similar, sectors and locations? different, or even opposing goals is just another dimension of the analysis.’ Woods and Gronn (2009) (building on from an Distributed leadership and earlier article by Woods 2004) go further to suggest related concepts that many current manifestations of DL within schools and other organizations suffer from a ‘demo- While I was drafting this paper for publication, it was cratic deficit’ in that they stop short of advocating suggested that it might be useful to compile a com- the principles of ‘self-governance, protection from parative table of terms associated with DL that could arbitrary power, and legitimacy grounded in consent’ be used as a source for distinguishing between them. (p. 433) that form the basis of democracy, in favour

© 2011 The Author International Journal of Management Reviews © 2011 British Academy of Management and Blackwell Publishing Ltd. Distributed Leadership in Organizations 257 of a rationale based on measures of efficiency and because it has normative, representational and effectiveness. empirical power within this context – both describing Thus, while there are clear parallels between many leadership as it is and as it should be. of these ideas, and a fair degree of similarity in their However, while the commonalities between DL theoretical and historical origins, like much else in and SL perspectives are evident, and may add the field of leadership they remain ‘essentially con- strength to the argument against leader-centric rep- tested concepts’ (Grint 2005). In a review of the resentations, there are some potential dangers of literature on DL (including the associated concepts assuming too close a similarity. As Leithwood et al. of ‘delegated’, ‘democratic’, and ‘dispersed’ leader- (2006) suggest, although there is clearly a degree of ship), though, Bennett et al. (2003, p. 7) did manage overlap between concepts of shared, collaborative, to identify three premises that seem to be shared by democratic and participative leadership, this does not most authors: mean that all forms are equal and/or equivalent, or that everybody is a leader. Indeed, a tendency to 1. ‘Leadership is an emergent property of a group or conflate and/or group together similar perspectives network of interacting individuals’ leads to the possibility that DL may become a label 2. ‘There is openness to the boundaries of leader- for all attempts to share or devolve leadership to ship’ others, which may, in turn, undermine its conceptual 3. ‘Varieties of expertise are distributed across the rigour and subsequent utility as a framework for many, not the few.’ investigating leadership practice (Harris 2005). Such criteria are common across much recent theor- In an attempt to set the record straight, Spillane izing on leadership as a relational process (see Uhl- and Diamond (2007a, pp. 149–152) dispel four Bien 2006 for a review), as well as SL theory (Pearce common ‘myths of distributed leadership’ as follows: et al. 2008). In the introduction to their book Shared (1) that DL is a blueprint for leadership and manage- Leadership, for example, Pearce and Conger (2003b, ment; (2) that DL negates the role of school princi- p. 1) define SL as: pals (or CEOs elsewhere); (3) that from a distributed perspective, everyone is a leader; and (4) that DL is a dynamic, interactive influence process among only about collaborative situations. Such assump- individuals in groups for which the objective is to tions, they argue, are fundamentally flawed and lead one another to the achievement of group or diminish our ability to get behind the dynamics of organizational goals or both. This influence process leadership practice. In the same way as leader-centric often involves peer, or lateral, influence and at other perspectives reduce our ability to recognize factors times involves upward or downward hierarchical beyond the individual, a poorly defined concept of influence. DL may render it difficult to differentiate the specific Similar distinctions between ‘vertical’ and ‘horizon- contributions of particular actors and/or aspects of tal’ leadership are made by researchers within the the situation. field of DL (e.g. Bolden et al. 2009; Collinson and Collinson 2009; Gronn 2009) and both sets of literature draw attention to the need to recognize informal, Patterns and outcomes of DL emergent and collective acts of influence as well as those instigated by people in formal positions of From the account above, it would seem that one of authority. the main distinctions between DL and more tradi- The perceived need for both DL and SL theory tional approaches is its attempt to offer a systemic similarly arise in response to a similar set of practical perspective on leadership rather than positioning challenges as summarized by Leithwood et al. itself as a distinct theory per se. Within any given (2009a, p. xvii) in the preface to their book Distrib- situation, DL may comprise a number of different uted Leadership according to the Evidence, where (and possibly competing) ‘configurations’4 (Gronn they suggest that DL responds to the ‘decidedly unheroic’ leadership experienced by many within 4 schools, as well as the ‘more complex mission facing Gronn (2009) proposes the idea of ‘leadership configuration’ – ‘a pattern or an arrangement of practice’ (p. 383) – in schools’ and a growing appreciation of the impor- an attempt to overcome the tendency to define DL in nor- tance of ‘informal’ leadership. Harris (2009) con- mative or ideological terms. From this perspective, he argues cludes that DL has been well received within schools ‘in any organization in which there may be evidence of

Table2.FrameworksofDL

Gronn (2002) Leithwood et al. (2006) MacBeath et al. (2004) Spillane (2006)

• Spontaneous collaboration: • Planful alignment: where, • Formal distribution: where • Collaborated distribution: where groups of individuals following consultation, leadership is intentionally where two or more with differing skills, resources and delegated or devolved. individuals work together in knowledge and/or responsibilities are • Pragmatic distribution: time and place to execute capabilities come together to deliberately distributed to where leadership roles and the same leadership routine. complete a particular task/ those individuals and/or responsibilities are • Collective distribution: project and then disband. groups best placed to lead a negotiated and divided where two or more • Intuitive working relations: particular function or task. between different actors. individuals work separately where two or more • Spontaneous alignment: • Strategic distribution: where but interdependently to enact individuals develop close where leadership tasks and new people, with particular a leadership routine. working relations over time functions are distributed in skills, knowledge and/or • Coordinated distribution: until ‘leadership is manifest an unplanned way yet, ‘tacit access to resources, are where two or more in the shared role space and intuitive decisions about brought in to meet a individuals work in encompassed by their who should perform which particular leadership need. sequence in order to relationship’ (p. 657). leadership functions result in • Incremental distribution: complete a leadership • Institutionalized practice: a fortuitous alignment of where people acquire routine. where enduring functions across leadership leadership responsibilities organizational structures sources’ (Harris et al., 2007, progressively as they gain (e.g. committees and teams) p. 344). experience. are put in place to facilitate • Spontaneous misalignment: • Opportunistic distribution: collaboration between where, as above, leadership is where people willingly take individuals. distributed in an unplanned on additional responsibilities manner, yet in this case the over and above those outcome is less fortuitous and typically required for their there is a misalignment of job in a relatively ad hoc leadership activities. manner. • Anarchic misalignment: where • Cultural distribution: where leaders pursue their own goals leadership is naturally independently of one another assumed by members of an and there is ‘active rejection, organization/group and on the shared organically between part of some or many individuals. organizational leaders, of inﬂuence from others about what they should be doing in their own sphere of inﬂuence’ (p. 344).

2009) of leadership practice – the cumulative inter- More important, what is likely to be most salient is action of which, give rise to outcomes. As Spillane not the fact that leadership is distributed but how (2006, pp. 102–103) suggests: leadership is distributed [. . .] A distributed perspective on leadership can coexist with and be used beneficially to explore hierarchical and top-down persons and units leading, that configuration is simply one of “leadership”, unqualified and unembellished, the practice of leadership approaches. which happens to be shaped in contextualized ways’ In order to more clearly articulate the various ways in (p. 390). Taking such a perspective on leadership, he suggests, carries two major implications: firstly, that researchers which DL can occur, a number of authors have devel- would be advised to identify and map the multiple hybrid oped taxonomies, four of which are summarized in forms of leadership that occur within a particular Table 2. organization/context longitudinally over time (rather than Although each of these frameworks is derived decontextualizing them and focusing only on particular from research in schools, it is possible to consider forms); and secondly, that we should avoid labelling different forms of leadership in a way that invokes normative how they might be applied in other contexts. Those comparisons between them (focusing instead on the under- suggested by Gronn (2002) and Spillane (2006), for lying bases upon which leadership is founded). example, focus on the interpersonal dynamics of DL

© 2011 The Author International Journal of Management Reviews © 2011 British Academy of Management and Blackwell Publishing Ltd. Distributed Leadership in Organizations 259 and the various ways in which people can collaborate ership can have a negative effect on team perform- to achieve shared outcomes (processes that are likely ance, including a ‘dispersion’ of responsibility to occur in most organizations). Those by MacBeath (Festinger et al. 1950; Heinicke and Bales 1953), a et al. (2004) and Leithwood et al. (2006) focus more reduced sense of stability and security (Melnick explicitly on different forms of DL and, while 1982) and boundary management issues (Storey perhaps more specific to a school context, may well 2004; Timperley 2005). be observable elsewhere. Each of these frameworks Within the field of school leadership, where the indicates a degree of variation in the extent to which majority of DL research has been conducted, May- DL is ‘institutionalized’ within working practices as rowetz (2008, p. 424) proposes ‘there is no strong part of the overall ‘culture’ of the organization, and link between distributed leadership and two primary the extent to which this may be instigated deliber- goals of the educational leadership field: school ately in a coordinated manner. While Gronn, Spillane improvement and leadership development’. Similarly and MacBeath do not suggest that one or more forms York-Barr and Duke (2004) concludes that there is of distribution are more effective or desirable than the little evidence of a direct impact of DL on student others – in that they each constitute part of the overall learning outcomes. landscape of leadership practice – Leithwood and A number of other studies (e.g. Harris and Muijs colleagues suggest that certain forms are more likely 2004; Hulpia and Devos 2009; Leithwood and Jantzi to contribute towards organizational productivity, as 2000; Silins and Mulford 2002), however, do indicate follows: a positive relationship between DL and significant aspects of school performance. Day et al. (2007, • Planful and spontaneous alignment are most likely p. 17) concluded that ‘substantial leadership distri- to contribute towards short-term organizational bution was very important to a school’s success in productivity. improving pupil outcomes’, albeit that this was a • Planful alignment is most likely to contribute mediated relationship, whereby DL ‘was positively towards long-term organizational productivity. correlated to the conditions within the organization, • Spontaneous misalignment and anarchic align- including staff morale, which in turn impacted posi- ment are likely to have a negative effect on short- tively upon student behaviour and student learning and long-term organizational productivity. outcomes’ (Day et al. 2009, cited in Harris 2009, It is also suggested that organizational members p. 15). Harris concludes: associated with the fourth configuration ‘anarchic The empirical evidence about distributed leadership misalignment’ are likely to be far more resistant to and organizational development is encouraging but the implementation of ‘planful alignment’ than far from conclusive. We need to know much more would those in contexts of spontaneous alignment or about the barriers, unintended consequences and misalignment. limitations of distributed leadership before offering The work by Leithwood and colleagues, cited any advice or prescription. We also need to know above, gives some indication of the potential benefit the limitations and pitfalls as well as the opportu- of a carefully implemented approach to DL, as well nities and potential of this model of leadership prac- as the dangers of a poorly conceived approach. It is tice. (Harris 2009, p. 18) interesting in that it indicates that ‘distributing’ lead- A very similar conclusion is reached by Leithwood ership per se is not necessarily beneficial, but that et al. (2009c) and leads them to suggest that, without what is important is how leadership is distributed. the ‘more nuanced appreciation of the anatomy of This may help account for the somewhat inconsis- distributed leadership’ (p. 280) which has developed tent evidence on the impact of DL on organizational only recently, it would be unrealistic to expect to find performance as illustrated in a recent review by a significant relationship between DL and perform- Harris (2009). While she identified a number of ance outcomes. studies indicating a positive relationship between Together, these authors, and the work upon which DL and organizational change (e.g. Graetz 2000; they draw, suggest that a key focus for future Iandoli and Zollo 2008), teacher leadership (e.g. research is exploring how particular configurations Little 1990; Rosenholtz 1989) and professional of DL contribute towards, or inhibit, organizational learning communities (e.g. Louis and Marks 1998; performance. A focus on the how of leadership dis- Morrisey 2000; Stoll and Louis 2007); there was tribution, however, is only part of the story. Other equally evidence to suggest that distribution of lead- important questions include why leadership is dis-

tributed, who controls this distribution and what The context of DL (if anything) is being distributed? Day et al. (2009, p. 14) suggest that in schools, for example: Power and influence. A number of authors (e.g. Brown et al. 2000; Gordon 2010; Gronn 2009; It is often some form of external pressure that Hartley 2009; Hatcher 2005) have suggested that prompts efforts to distribute leadership more much current theory and research on DL takes insuf- broadly, for example, pressure to improve dis- ficient consideration of the dynamics of power and appointing school performance, introduction of new influence in which it is situated. Within schools, for policies and programmes requiring new teaching example, Hatcher (2005) concludes, that while lead- and learning capacities. Greater ‘distribution of leadership’ outside of those in formally established ership may be ‘distributed’, power often is not. The roles usually depends on quite intentional inter- notion of DL may be invoked by senior managers to vention on the part of those in formal leadership encourage engagement and participation in organ- roles. izational activities while masking substantial imbalances in access to resources and sources of power. From this perspective, the implementation of DL Gunter and Rayner (2007b), similarly propose that cannot be considered as politically neutral. It is moti- ‘that the labelling of Headteachers as managing vated by a series of expectations, assumptions and directors, chief executives, and more recently as agendas which, while at face value may appear rea- school leaders, combined with training and a require- sonable enough, may actually serve to legitimize and ment to implement reforms, is central to a form of reinforce the domination of particular individuals modernization that is reworking professionality as and groups over others. generic and business orientated’. Such shifts are Furthermore, by assuming that leadership is some- common across much public sector reform and thing that can be distributed and mapped in various require the merging of managerial and professional ways we risk reinforcing an essentialist understand- identities in ways that may prove problematic for ing of the concept which is one of the very things that some occupations (see Gosling et al. 2009 for a DL offers the potential of overcoming (i.e. the tran- similar argument in higher education). sition from an individualistic to a social process view The significance of ‘leadership’ as a vehicle for of leadership). public sector reform is now relatively widely recog- These are ideas that are explored further in the nized (see, for example, Brooks 2000; Currie et al. next section. 2005; Deem et al. 2007; Gleeson and Knights 2008; Hartley and Allison 2000) yet is not always acknow- ledged within research on DL in these contexts. What might we still be missing? Where it has been (e.g. Day 2005; Gunter and Rayner 2007a; Hargreaves and Fink 2008; Hartley and The notion of DL, as illustrated in this paper, has Allison 2000; Torrance 2009), authors have high- clearly been influential in shaping how leadership lighted a number of challenges and inconsistencies in has been conceived and investigated over the past how the notion of DL is framed by such bodies. In a decade. As Harris and Spillane (2008, p. 33) recent study within schools, for example, Currie propose: ‘it is a way of getting under the skin of et al. (2009, p. 1735) illustrate how ‘competing insti- leadership practice, of seeing leadership practice tutional forces simultaneously foster and stymie the differently and illuminating the possibilities for adoption of distributed leadership’ such that ‘school organizational transformation’. While drawing our principals find themselves in a classic Catch-22 situ- attention to some previously neglected aspects of ation, which they resolve by enacting a weak form of leadership practice, however, it is possible that the distributed leadership’. ‘Ironically’, they argue, ‘the concept of DL has obscured other important aspects implementation of distributed leadership is the most of organizational life. In this section, I consider a difficult in the schools located in socially deprived number of areas worthy of further consideration, areas, that is, the very context where policy-makers beginning with three aspects of the context of expect distributed leadership to make the most leadership that tend to be under-theorized impact’. and/or explored in accounts of DL, followed by To appreciate why organizations (and other social three methodological/developmental challenges that groups) operate in the ways they do, it is important to demand further attention. remain alert to the dynamics of power and influence,

© 2011 The Author International Journal of Management Reviews © 2011 British Academy of Management and Blackwell Publishing Ltd. Distributed Leadership in Organizations 261 and how they enable or constrain particular forms of the role of the principal, as do most of those in engagement in leadership practice. Gordon (2010) Leithwood et al. (2009a). Such a situation severely provides a graphic illustration of how this can be limits opportunities for recognizing the contribution done in his analysis of the implementation of a ‘dis- of informal leaders and the manner in which situ- persed leadership’ initiative within an Australian ational factors (physical, social and cultural) impact police department. Through the application of a upon leadership. framework of antecedent forms of power (taken for McCrimmon (2005) proposes, for example, that granted realities, historical delineation of relation- ‘thought leadership’ within networks and partnerships, historical decision legitimacy, the ordering of ships is a significant and relatively unexplored statements, and boundaries of discursive actions), he dimension of DL that deserves investigation. Pye and demonstrates how ‘rather than dispersing power and Knight (2005) go further to suggest that the network enhancing democracy, antecedent forms of power itself offers an important level of analysis which continue to legitimize domination on behalf of those often goes neglected yet is essential to our under- groups and individuals (senior officers and detec- standing of how leadership occurs within and tives) who previously held formal positions of between organizations. power’ (p. 283). By way of conclusion, he suggests Furthermore, while there is some literature explor- that ‘leadership implies power as much as power ing DL outside the education sector (e.g. Ensley implies leadership, and thus leadership innocent of et al. 2006; Mehra et al. 2006; Nonaka and Toyama power is leadership ignorantly normalized: Power is 2002) this remains rare and, in order to test the valid- always implicated with the discourse and practices of ity and utility of a distributed perspective more leadership’. According to Gordon, such analyses are widely, further research is required. In particular, woefully absent within the literature on dispersed work that enables comparison of the relative desir- leadership (including DL and SL), which ‘adopts a ability and/or appropriateness of terms such as normative, apolitical approach to power’ (pp. ‘distributed’, ‘shared’ or ‘dispersed’ leadership in 281–282). different contexts could be helpful in clarifying To truly acknowledge the dynamics of power and whether differences in terminology are purely rhet- influence that permeate our organizations and soci- orical or whether they point to more fundamental eties, we need to pay more attention to issues such as differences in how leadership is accomplished. competition between leaders (Storey 2004), micropolitics (Bjork and Blase 2009) and the rhetoric of Ethics and diversity. I have already discussed the partnerships (Lumby 2009). warning by Woods and Gronn (2009) that important debates about the value of democracy are often Organizational boundaries and context. A further skirted over in accounts of DL, yet other authors limitation within much work on DL is the tendency highlight a number of additional challenges. Sugrue to confine studies within organizational boundaries. (2009), for example, argues that, in focusing on the Thus, for example, studies of school leadership tend collective aspects of leadership, we may ‘short to explore the contributions of various actors within change’ and ‘emasculate’ the ‘everyday heroes and the school and, perhaps, the influence of particular heroines’ whose essential contribution may go un- stakeholders such as pupils, parents and school gov- noticed. Rippin (2007) takes a feminist informed per- ernors. But they rarely, if ever, consider how leader- spective to suggest the contributions of individuals ship practices within one school may impact upon (particularly females) may remain largely invisible those within another, and/or how their effects may through the manner in which gendered work is con- ripple through the community to impact upon local structed and interpreted. Grint (2010) suggests that business, healthcare, policing, etc. Furthermore, the DL ignores the inherently ‘sacred’ nature of leader- temporal dimensions of how leadership patterns ship work, and the importance of separation, sacri- experienced at school frame and shape expectations fice and silencing of and by ‘leaders’. And Sinclair of leadership in colleges, universities and workplaces (1992) highlights the potentially hegemonic nature remain well beyond the scope of current enquiries. of a team-based ideology. Together, these sources There is also a tendency for most research on DL indicate that the shift to a more collective or distrib- to focus on the holders of formal positions. For uted representation of leadership does not necessar- example, each of the six cases of ‘DL’ presented in ily have a beneficial effect for those people involved Spillane and Diamond (2007a) focus particularly on with it.

With regard to the issue of diversity, while DL up interpreting other factors as leadership (e.g. offers the promise of a more inclusive perspective on Lakomski 2008; Washbush 2005). From this per- leadership, which incorporates a range of views and spective, searching for the ‘essence’ of leadership contributions, much of the work fails to take a cross/ and/or direct evidence of the impact of leadership, multicultural perspective (for an exception, see ‘distributed’ or otherwise, is meaningless. Instead, McIntyre 2003; Yeung et al. 2006) and how to truly we should focus our attention on the processes by recognize and respond to the needs and expectations which certain things, and not others, are categorized of end-users, e.g. student voice in schools (see as leadership (Kelly 2008; Pondy 1978). Such a per- Menon 2005; Mitra 2005) or even cultural prefer- spective would give a fuller appreciation of the role ences for top-down leadership (Goldstein 2004). of language in the accomplishment of leadership (Fairhurst 2007; Grint 2005; Gronn 1983), as well as the importance of rhetoric and discourse in shaping Methodological and developmental challenges leadership expectations (Fitzgerald and Gunter 2007; Ontology. A further important issue which may Hatcher 2005; Maxcy and Nguyen 2006). well be elided by the current focus on DL is the ontological status of leadership itself. While DL Research methods. Following on from the account rejects the notion that leadership resides within indi- given above about the need for increased recognition viduals in preference for a relational ontology, it of the primacy of contextual factors in shaping both stops short of challenging the underlying building how leadership practice occurs and is understood, blocks of leadership theory or whether, indeed, lead- considerable care needs to be given to the manner in ership can (or should) be conceived of as a distinct which DL research is conducted and interpreted. As concept at all. With regard to the first of these issues, mentioned previously, much current research focuses Drath et al. (2008, p. 635) propose that the vast predominantly on the testimonies of key institutional majority of leadership theory, including DL, is actors in formal leadership/management roles. While ‘unified and framed by an underlying ontology that is such people clearly play an important role in the virtually beyond question within this field’. This enactment and interpretation of leadership, their per- ontology, they propose, is one of the ‘tripod’: ‘an spectives may differ in significant ways from those expression of commitment to the entities (leaders, elsewhere within and/or outside the organization. followers, common goals) that are essential and Understanding and exploring these apparent tensions indispensable to leadership and about which any and contradictions may well shed light on the discur- theory of leadership must therefore speak’ (ibid, sive processes through which leadership occurs p. 635). These elements are clearly evident with DL and the importance of different ‘social identities’ theories such as that proposed by Spillane et al. (Haslam et al. 2003) in shaping attitudes and (2004) and, while they offer an explanation as to why behaviours. they retain the terminology of ‘leaders’ and ‘follow- Other authors suggest that studies should focus ers’ (see, for example, Spillane and Diamond 2007b, more directly on the impact of DL practices on pp. 8–9) this distinction remains problematic for any key organizational outcomes. Within schools, for representation that endeavours to escape assigning example, Timperley (2005) and Robinson (2009) ontological primacy to individual agents (Hartley suggest that student learning is the key variable and 2009). that we should endeavour to capture the ways in With regard to the second issue, DL theory tends which different forms of leadership contribute not to question the existence of leadership as a dis- towards this. tinct concept, despite the growing literature on the Youngs (2009) suggests that the field of DL is social construction of leadership (see Fairhurst and drifting towards a somewhat uncritical position and Grant 2010, for a recent review) and recognition that highlights four key areas of concern: (1) a lack of ‘leadership is an emergent property of a group or critique against policy; (2) an under-emphasis of network of interacting individuals’ (Bennett et al. historical precedents; (3) ignorance of parallel 2003, p. 7). Authors from a more critical perspective developments; and (4) a lack of attention to power suggest that ‘thinking about leadership needs to take relations. He calls for the development of a critical seriously the possibility of the non-existence of lead- perspective to complement and challenge the ership as a distinct phenomenon’ (Alvesson and descriptive and normative approaches which Sveningsson 2003, p. 359) and that we may well end dominate the literature.

Methodologically, the evidence from this review Conclusions supports a shift in focus from simply studying how leadership is ‘distributed’ to a contextually situated In this paper I have given an overview of the field of exploration of how distributed and focused forms of DL, how it has developed and areas for further con- leadership interact with one another within a ‘hybrid sideration. It has been demonstrated that the concept configuration’ of practice (Gronn 2008b, 2009, of DL bears many similarities to notions such as 2010). Such an approach would require detailed eth- ‘shared’, ‘collective’, ‘collaborative’, ‘emergent’ and nographic studies of leadership practices and dis- ‘co-’ leadership and has some common theoretical courses in situ, as well as a multi-level approach to and practical origins. Despite this, there remain some research (Yammarino and Dansereau 2008). Greater significant differences in how these concepts are uti- use of critical discourse analysis could also be impor- lized and their relative prevalence within different tant in shedding light on underlying dynamics of parts of the literature. There are also some key areas power and influence and the rhetorical significance of debate, such as the extent to which DL is neces- of DL terminology (Gordon 2010; Gosling et al. sarily inclusive or democratic; the degree to which it 2009). should be taken as a framework for improving leadership practice or simply describing it; and the Development, reward and recognition of leaders. degree to which the concept of DL is part of a wider Finally this review illustrates the manner in socio-political movement aiming to reform public which, from a distributed perspective, the practice services and produce efficiency savings (Hargreaves of leadership is integrally bound up with the and Fink 2008; Hartley 2007). wider system in which it occurs. To this extent, From a review of DL literature in schools May- structural and situational factors are ‘constitutive’ rowetz (2008, p. 424) proposes that ‘there are four elements of leadership practice (Spillane et al. common usages of the term DL, which include the 2004), as are the contributions of multiple actors, original descriptive theoretical lens and three pre- not just those in formal managerial roles. Such scriptions for how sharing leadership in schools can insights have important implications for the manner improve practice’. He suggests that ‘each usage has in which leadership is recognized, rewarded and its strengths and weaknesses, though two of the developed within organizations yet fail to have prescriptive usages are actually contradicted by seriously impacted upon much practice in this empirical research’. He concludes that ‘while not area. dismissing any particular definition, the author James et al. (2007) and Ross et al. (2005a,b) encourages those who use the descriptive definition outline the need for a more systemic approach to to focus more on making connections to leadership development that situates this activity as school improvement and leadership development part of a wider change process. Busher (2005), Loder [...and...]those who use the prescriptive defini- and Spillane (2005) and others draw attention to the tions to use theoretically and empirically grounded importance of ‘identity work’ (Sveningsson and research frames and offer suggestions about how to Alvesson 2003) in enabling people to work through link research on distributed leadership to the practi- potentially competing identities in relation to leader- cal concerns of the field’. From the evidence in the ship work. Louis et al. (2009) and Simkins (2005) current paper, I would endorse these views to suggest highlight the importance of collective sense-making that a purely descriptive approach is of limited use in and the establishment of trust in developing effective enhancing leadership practice, while a normative models of DL. approach may inadvertently end up promoting in- Together, this literature poses some serious chal- appropriate, ineffective and potentially unethical lenges to traditional management and leadership practices. Like Youngs (2009) and Gordon (2010), I development processes and calls for far greater would advocate a more critical perspective which investment in the development of interpersonal net- facilitates reflection on the purpose(s) and discursive works and shared understandings both within and mechanisms of leadership and an awareness of the beyond organizations (what Day (2000) refers to as dynamics of power and influence in shaping what ‘leadership development’ as opposed to ‘leader happens within and outside organizations. Further- development’). Elaboration on how this might be more, the rhetorical significance of DL in (re)con- done effectively remains an important area for structing leader identities and mobilizing collective further research. engagement should not be underestimated, as well as

© 2011 The Author International Journal of Management Reviews © 2011 British Academy of Management and Blackwell Publishing Ltd. 264 R. Bolden the inherent paradoxes and tensions of much leader- Alvesson, M. and Sveningsson, S. (2003). The great dis- ship practice (Bolden et al. 2008; Gosling et al. appearing act: difficulties in doing ‘leadership’. Leader- 2009). ship Quarterly, 14, pp. 359–381. Gronn’s notion of ‘hybrid configurations’ of lead- Astley, G. (1985). Administrative science as socially con- ership (Gronn 2008b, 2009, 2010) offers a promising structed truth. Administrative Science Quarterly, 30, pp. 497–513. approach to moving beyond overly simplistic or aspi- Badaracco, J.L. (2001). We don’t need another hero. rational views of DL which may help to shed light on Harvard Business Review, 79(8), pp. 120–126. the important balance between individual, collective Barker, R. (2001). The nature of leadership. Human Rela- and situational aspects of leadership practice and, tions, 54, pp. 469–494. importantly, when and why particular configurations Barnard, C.I. (1968). Functions of the Executive. Cam- are more effective and/or desirable than others. As bridge, MA: Harvard University Press. Pearce (2004, p. 55) suggests: Barry, D. (1991). Managing the bossless team: lessons in distributed leadership. Organizational Dynamics, 20, pp. The issue is not vertical leadership or shared lead- 31–47. ership. Rather the issues are: (1) when is leadership Bass, B.M. (1985). Leadership and Performance beyond most appropriately shared? (2) How does one Expectations. New York: Free Press. develop shared leadership? And (3) how does one Beck, A.P. (1981). A study of group phase development and utilize both vertical and shared leadership to lever- emergent leadership. Group Dynamics: Theory, Research, age the capabilities of knowledge workers? and Practice, 5, pp. 48–54. The key contribution of DL, it would seem, is not in Beck, A.P. and Peters, L. (1981). The research evidence for offering a replacement for other accounts, but in distributed leadership in therapy groups. International Journal of Group Psychotherapy, 31, pp. 43–71. enabling the recognition of a variety of forms of Becker, H. and Useem, R.H. (1942). Sociological analysis of leadership in a more integrated and systemic manner. the dyad. American Sociological Review, 7, pp. 13– The potential ability of DL to achieve this, however, 26. is somewhat limited by its restriction to particular Benne, K.D. and Sheats, P.(1948). Functional roles of group contexts and locations, which leaves it somewhat members. Journal of Social Issues, 4, pp. 41–49. detached from the wider leadership literature,5 with Bennett, N., Wise, C., Woods, P.A. and Harvey, J.A. (2003). the apparent juxtaposition with ‘shared leadership’ a Distributed Leadership. Nottingham: National College of particular area for concern. School Leadership. To conclude, however, to be truly successful and to Bjork, L.G. and Blase, J. (2009). The micropolitics of school achieve the impact that it promises, the concept of district decentralization. Educational Assessment, Evalu- DL really needs to connect in a meaningful way with ation and Accountability, 21, pp. 195–208. Bolden, R., Petrov, G. and Gosling, J. (2008). Tensions in the experiences and aspirations of leadership practi- higher education leadership: towards a multi-level model tioners (Harris and Spillane 2008), as well as explic- of leadership practice. Higher Education Quarterly, 62, itly recognizing the inherently political nature of pp. 358–376. leadership within organizations and imbalances in Bolden, R., Petrov, G. and Gosling, J. (2009). Distributed the distribution of power and influence (Gordon leadership in higher education: rhetoric and reality. Edu- 2010; Woods and Gronn 2009). Only then will it be cational Management, Administration and Leadership, able to move beyond adolescence to maturity. 37, pp. 257–277. Brooks, J. (2000). Labour’s modernization of local govern- ment. Public Administration Review, 78, pp. 593–612. References Brown, J.S. and Duguid, P. (1991). Organizational learning and communities-of-practice: toward a unified view of Alvesson, M. (1996). Leadership studies: from procedure working, learning, and innovation. Organizational and abstraction to reflexivity and situation. Leadership Science, 2, pp. 40–57. Quarterly, 7(4), pp. 455–485. Brown, M., Rutherford, D. and Boyle, B. (2000). Leadership for school improvement: the role of the head of department in UK secondary schools. School Effectiveness and 5Youngs(2009, p. 385), for example, remarks ‘as far as I can ascertain the works of Peter Gronn, and to a lesser extent School Improvement, 11, pp. 237–258. Kenneth Leithwood, usually incorporate links to the wider Brown, M.E. and Gioia, D.A. (2002). Making things leadership field. The generic leadership journal, Leadership click: distributive leadership in an online division of an Quarterly, on the other hand, only tends to highlight Gronn’s offline organization. Leadership Quarterly, 13, pp. 397– theorizing of distributed leadership.’ 419.

Brown, M.M. and Hosking, D.D. (1986). Distributed lead- Fairhurst, G.T. (2007). Discursive Leadership: In Conversa- ership and skilled performance as successful organization tion with Leadership Psychology. London: Sage. in social movements. Human Relations, 39, pp. 65–79. Fairhurst, G.T. and Grant, D. (2010). The social construction Busher, H. (2005). Being a middle leader: exploring profes- of leadership: a sailing guide. Management Communica- sional identities. School Leadership and Management, 25, tion Quarterly, 24, pp. 171–210. pp. 137–153. Festinger, L., Schacter, S. and Back, K.W. (1950). Social Collinson, D. and Collinson, M. (2009). ‘Blended leader- Pressure in Informal Groups: A Study of Human Factors ship’: employee perspectives on effective leadership in the in Housing. New York: Harper. UK further education sector. Leadership, 5, pp. 365–380. Fitzgerald, T. and Gunter, M. (2007). Teacher Leadership: A Currie, G., Boyett, I. and Suhomlinova, O. (2005). Transfor- New Myth for Our Time? Chicago, IL: AERA. mational leadership within secondary schools in England: Follett, M.P. (1942/2003). Dynamic Administration: The a panacea for organisational ills? Public Administration, Collected Papers of Mary Parker Follett. London: 83, pp. 265–296. Routledge. Currie, G., Lockett, A. and Suhomlinova, O. (2009). The French, J.J.R.P. and Snyder, R. (1959). Leadership and inter- institutionalization of distributed leadership: a ‘Catch-22’ personal power. In Cartwright, D. (ed.), Studies in Social in English public services. Human Relations, 62, pp. Power. Ann Arbor, MI: Institute for Social Research, Uni- 1735–1761. versity of Michigan, pp. 118–149. Dahl, R.A. (1961). Who Governs? Democracy and Power in Gibb, C.A. (1954). Leadership. In Lindzey, G. (ed.), Hand- an American City. New Haven, CT: Yale University Press. book of Social Psychology, Vol. 2. Reading, MA: Day, C. (2005). The UK policy for school leadership: uneasy Addison-Wesley, pp. 877–917. transitions. In Bascia, N., Cumming, A., Datnow, A., Giddens, A. (1979). Central Problems in Social Theory. Leithwood, K. and Livingstone, D. (eds), International London: Macmillan. Handbook of Educational Policy. Dordrecht: Springer, Giddens, A. (1982). Proﬁles and Critiques in Social Theory. pp. 392–420. Berkeley, CA: University of California Press. Day, C. et al. (2007). The Impact of School Leadership on Gleeson, D. and Knights, D. (2008). Reluctant leaders: an Pupil Outcomes: Interim Report. Nottingham: National analysis of middle managers’ perceptions of leadership in College for School Leadership for Department for Chil- further education in England. Leadership, 4, pp. 49–72. dren, Schools and Families. Goldstein, J. (2004). Making sense of distributed leadership: Day, C. et al. (2009). The Impact of School Leadership on the case of peer assistance and review. Educational Pupil Outcomes: Final Report. Nottingham: Department Evaluation and Policy Analysis, 26, pp. 173–197. for Children, Schools and Families. Gordon, R.D. (2002). Conceptualizing leadership with Day, D.V. (2000). Leadership development: a review in respect to its historical contextual antecedents to power. context. Leadership Quarterly, 11, pp. 581–613. Leadership Quarterly, 13, pp. 151–167. Deem, R., Reed, M. and Hillyard, S. (2007). Knowledge, Gordon, R.D. (2010). Dispersed leadership: exploring the Higher Education, and the New Managerialism: The impact of antecedent forms of power using a communi- Changing Management of UK Universities. Oxford: cative framework. Management Communication Quar- Oxford University Press. terly, 24, pp. 260–287. Denis, J.L., Lamothe, L. and Langley, A. (2001). The Gosling, J., Bolden, R. and Petrov, G. (2009). Distributed dynamics of collective leadership and strategic change in leadership in higher education: what does it accomplish? pluralistic organizations. Academy of Management Leadership, 5, pp. 299–310. Journal, 44, pp. 809–837. Graetz, F. (2000). Strategic change leadership. Management Drath, W.H., McCauley, C.D., Palus, C.J., Van Velsor, E., Decisions, 38, pp. 550–562. O’Connor, P.M.G. and McGuire, J.B. (2008). Direction, Gregory, M. (1996). Developing effective college leadership alignment, commitment: toward a more integrative ontol- for the management of educational change. Leadership ogy of leadership. Leadership Quarterly, 19, pp. 635–653. and Organizational Development Journal, 17, pp. 46–51. Engeström, Y. (1999). Activity theory and individual and Grint, K. (2005). Leadership: Limits and Possibilities. Bas- social transformation. In Engeström, Y., Miettinen, R. and ingstoke: Palgrave Macmillan. Punamaki, R.L. (eds), Perspectives on Activity Theory. Grint, K. (2010). The sacred in leadership: separation, Cambridge: Cambridge University Press, pp. 19–38. sacriﬁce and silence. Organization Studies, 31, pp. 89– Ensley, M.D., Hmieleski, K.M. and Pearce, C.L. (2006). The 107. importance of vertical and shared leadership within new Gronn, P. (1983). Talk as the work: the accomplishment of venture top management teams: implications for the per- school administration. Administrative Science Quarterly, formance of startups. Leadership Quarterly, 17, pp. 217– 28, pp. 1–21. 231. Gronn, P. (2000). Distributed properties: a new architecture Etzioni, A. (1965). Dual leadership in complex organiza- for leadership. Educational Management Administration tions. American Sociological Review, 30, pp. 688–698. & Leadership, 28, pp. 317–338.

Gronn, P.(2002). Distributed leadership as a unit of analysis. Hatcher, R. (2005). The distribution of leadership and power Leadership Quarterly, 13, pp. 423–451. in schools. British Journal of Sociology of Education, 26, Gronn, P. (2006). The significance of distributed leadership. pp. 253–267. Educational Leadership Research, 7. http://slc.educ.ubc. Heenan, D. and Bennis, W. (1999). Co-Leaders: The Power ca/eJournal/Issue7/index7.html of Great Partnerships. Hoboken, NJ: Wiley. Gronn, P. (2008a). The future of distributed leadership. Heinicke, C.M. and Bales, R.F. (1953). Developmental Journal of Educational Administration, 46, pp. 141– trends in the structure of small groups. Sociometry, 16, 158. pp. 7–38. Gronn, P. (2008b). Hybrid leadership. In Leithwood, K., Hosking, D.M. (1988). Organising, leadership and skillful Mascall, B. and Strauss, T. (eds), Distributed Leadership process. Journal of Management Studies, 25, pp. 147– according to the Evidence. Abingdon: Routledge, pp. 166. 17–40. Hulpia, H. and Devos, G. (2009). Exploring the link between Gronn, P. (2009). Leadership configurations. Leadership, 5, distributed leadership and job satisfaction of school pp. 381–394. leaders. Educational Studies, 35, pp. 153–171. Gronn, P. (2010). Hybrid configurations of leadership. In Hutchins, E. (1995). Cognition in the Wild. Cambridge, MA: Bryman, A., Collinson, D., Grint, K., Jackson, B. and MIT Press. Uhl-Bien, M. (eds), Sage Handbook of Leadership. Iandoli, L. and Zollo, G. (2008). Organisational Cognition London: Sage, pp. 435–452. and Learning. New York: Idea Group Incorporated. Gunter, H. and Rayner, S. (2007a). Modernizing the school James, K.T., Mann, J. and Creasy, J. (2007). Leaders as lead workforce in England: challenging transformation and learners: a case example of facilitating collaborative lead- leadership? Leadership, 3, pp. 47–64. ership learning for school leaders. Management Learning, Gunter, H.M. and Rayner, S. (2007b). Remodeling head- 38, pp. 79–94. teachers in England: is it the end of educational leader- Jaques, E. (1989). Requisite Organization: The CEO’s ship? International Electronic Journal for Leadership in Guide to Creative Structure and Leadership. Arlington, Learning, 11. Available at http://www.ucalgary.ca/iejll/ VA: Cason Hall. vol11/rayner Jermier, J.M. and Kerr, S. (1997). Substitutes for leadership: Hargreaves, A. and Fink, D. (2008). Distributed leadership: their meaning and measurement – contextual recollec- democracy or delivery? Journal of Educational Adminis- tions and current observations. Leadership Quarterly, 8, tration, 46, pp. 229–240. pp. 95–101. Harris, A. (2005). Leading or misleading? Distributed lead- Katz, D. and Kahn, R.L. (1966). The Social Psychology of ership and school improvement. Journal of Curriculum Organizations. New York: Wiley. Studies, 37, pp. 255–265. Katz, R.L. and Kahn, D. (1978). The Social Psychology of Harris, A. (2007). Distributed leadership: conceptual confu- Organizing, 2nd edn. New York: Wiley. sion and empirical reticence. International Journal of Kelly, S. (2008). Leadership: a categorical mistake? Human Leadership in Education, 10(3), pp. 315–325. Relations, 61, pp. 763–782. Harris, A. (2009). Distributed leadership: what we know. In Kerr, S. and Jermier, J.M. (1978). Substitutes for leadership: Harris, A. (ed.), Distributed Leadership: Different Per- their meaning and measurement. Organizational Behav- spectives. Dordrecht: Springer, pp. 11–21. ior and Human Performance, 22, pp. 375–419. Harris, A. and Muijs, D. (2004). Improving Schools through Lakomski, G. (2008). Functionally adequate but causally Teacher Leadership. London: Oxford University Press. idle: w(h)ither distributed leadership? Journal of Educa- Harris, A. and Spillane, J. (2008). Distributed leadership tional Administration, 46, pp. 159–171. through the looking glass. Management in Education, 22, Latour, B. (1987). Science in Action: How to Follow Scien- pp. 31–34. tists and Engineers through Society. Philadelphia, PA: Hartley, D. (2007). The emergence of distributed leadership Open University Press. in education: why now? British Journal of Educational Lave, J. and Wenger, E. (1991). Situated Learning: Legiti- Studies, 55, pp. 202–214. mate Peripheral Participation. New York: Cambridge Hartley, D. (2009). Education policy, distributed leadership University Press. and socio-cultural theory. Educational Review, 61, pp. Leithwood, K. and Jantzi, D. (2000). The effects of 139–150. different sources of leadership on student engagement Hartley, J. and Allison, M. (2000). The role of leadership in in school. In Riley, K. and Louis, K. (eds), Leadership the modernization and improvement of public services. for Change and School Reform. London: Routledge, Public Money and Management, 20, pp. 35–40. pp. 50–66. Haslam, S.A., Postmes, T. and Ellemers, N. (2003). Leithwood, K., Day, C., Sammons, P., Harris, A. and More than a metaphor: organizational identity makes Hopkins, D. (2006). Successful School Leadership: What organizational life possible. British Journal of Manage- It Is and How It Influences Pupil Learning. Nottingham: ment, 14, pp. 357–369. DfES Publications.

Leithwood, K., Mascall, B. and Strauss, T. (2009a). Distrib- McCrimmon, M. (2005). Thought leadership: a radical uted Leadership according to the Evidence. Abingdon: departure from traditional, positional leadership. Man- Routledge. agement Decision, 43, pp. 1064–1070. Leithwood, K., Mascall, B. and Strauss, T. (2009b). New McIntyre, J. (2003). Yeperenye dreaming in conceptual, geo- perspectives on an old idea: a short history of the old idea. graphical, and cyberspace: a participatory action research In Leithwood, K., Mascall, B. and Strauss, T. (eds), Dis- approach to address local governance within an Austra- tributed Leadership according to the Evidence. Abing- lian Indigenous housing association. Systemic Practice don: Routledge, pp. 1–14. and Action Research, 16, pp. 309–338. Leithwood, K., Mascall, B. and Strauss, T. (2009c). What we Mehra, A., Smith, B.R., Dixon, A.L. and Robertson, B. have learned and where we go from here. In Leithwood, (2006). Distributed leadership in teams: the network of K., Mascall, B. and Strauss, T. (eds), Distributed Leader- leadership perceptions and team performance. Leadership ship according to the Evidence. Abingdon: Routledge, pp. Quarterly, 17, pp. 232–245. 269–281. Meindl, J. (1995). The romance of leadership as a follower- Leithwood, K., Steinbach, R. and Ryan, S. (1997). Leader- centric theory: a social constructionist approach. Leader- ship and team learning in secondary schools. School ship Quarterly, 6, pp. 329–341. Leadership & Management, 17, pp. 303–326. Meindl, J.R., Ehrlich, S.B. and Dukerich, J.M. (1985). The Leont’ev, A.N. (1978). Activity, Consciousness and Person- romance of leadership. Administrative Science Quarterly, ality (tr. M.J. Hall). Englewood Cliffs, NJ: Prentice-Hall. 30, pp. 78–102. Leont’ev, A.N. (1981). Problems of the Development of the Melnick, M.J. (1982). Six obstacles to effective team per- Mind. Moscow: Progress Publishers. formance: some small group considerations. Journal of Lipman-Blumen, J. (1996). Connective Leadership: Manag- Sport Behavior, 5, pp. 114–123. ing in a Changing World. Oxford: Oxford University Menon, M.E. (2005). Students’ views regarding their par- Press. ticipation in university governance: implications for Little, J.W. (1990). The persistence of privacy: autonomy distributed leadership in higher education. Tertiary Edu- and initiative in teachers’ professional relations. Teachers cation and Management, 11, pp. 167–182. College Record, 91, pp. 509–536. Mitra, D.L. (2005). Adults advising youth: leading while Loder, T.L. and Spillane, J.P. (2005). Is a principal still a getting out of the way. Educational Administration Quar- teacher?: US women administrators’ accounts of role con- terly, 41, pp. 520–553. flict and role discontinuity. School Leadership and Man- Morrisey, M. (2000). Professional Learning Communities: agement, 25, pp. 263–279. An Ongoing Exploration. Austin, TX: Southwest Educa- Louis, K.S. and Marks, H. (1998). Does professional com- tional Development Laboratory. munity affect the classroom? Teachers’ work and student Nonaka, I. and Toyama, R. (2002). A firm as a dialectical work in restructuring schools. American Educational being: towards a dynamic theory of a firm. Industrial and Research Journal, 33, pp. 757–798. Corporate Change, 11, pp. 995–1009. Louis, K.S., Mayrowetz, D., Smiley, M. and Murphy, J. Northouse, P.G. (2007). Leadership: Theory and Practice, (2009). The role of sensemaking and trust in developing 4th edn. London: Sage. distributed leadership. In Harris, A. (ed.), Distributed O’Toole, J., Galbraith, J. and Lawler, E. (2003). When two Leadership. Dordrecht: Springer, pp. 157–180. (or more) heads are better than one: the promise and Lumby, J. (2009). Collective leadership of local school pitfalls of shared leadership. In Pearce, C.L. and Conger, systems: power, autonomy and ethics. Educational Man- J. (eds), Shared Leadership: Reframing the Hows and agement Administration & Leadership, 37, pp. 310–328. Whys of Leadership. London: Sage, pp. 250–267. MacBeath, J., Oduro, G.K.T. and Waterhouse, J. (2004). Oduro, G.K.T. (2004). Distributed leadership’ in schools: Distributed Leadership in Action: A Study of Current what English headteachers say about the ‘pull’ and ‘push’ Practice in Schools. Nottingham: National College for factors. Paper presented at the British Educational School Leadership. Research Association Annual Conference, University of Manz, C.C. and Sims, H.P., Jr (1993). Self-management as a Manchester, 16–18 September. substitute for leadership: a social learning perspective. Osborn, R.N. and Hunt, J.G. (2007). Leadership and the Academy of Management Review, 5, pp. 361–367. choice of order: complexity and hierarchical perspectives Maxcy, B.D. and Nguyen, T.S.o.n.T. (2006). The politics of near the edge of chaos. Leadership Quarterly, 18, pp. distributing leadership: reconsidering leadership distribu- 319–340. tion in two Texas elementary schools. Educational Policy, Parry, K. and Bryman, A. (2006). Leadership in organiza- 20, pp. 163–196. tions. In Clegg, S., Hardy, C., Lawrence, T. and Nord, W. Mayrowetz, D. (2008). Making sense of distributed leader- (eds), Sage Handbook of Organization Studies. London: ship: exploring the multiple usages of the concept in the Sage, pp. 447–468. field. Educational Administration Quarterly, 44, pp. 424– Pea, R.D. (1993). Practices of distributed intelligence and 435. designs for education. In Salomon, G. (ed.), Distributed

Cognition: Psychological and Educational Consider- Salomon, G. (1993). Distributed Cognition: Psychological ations. New York: Cambridge University Press, pp. and Educational Considerations. New York: Cambridge 47–87. University Press. Pearce, C. (2004). The future of leadership: combining ver- Schein, E.H. (1988). Organizational Psychology. Engle- tical and shared leadership to transform knowledge work. wood Cliffs, NJ: Prentice-Hall. Academy of Management Executive, 18, pp. 47–57. Senge, P.M. (1993). Transforming the practice of manage- Pearce, C. and Conger, J.A. (2003a). Shared Leadership: ment. Human Resource Development Quarterly, 4, pp. Reframing the Hows and Whys of Leadership. London: 5–32. Sage. Silins, H. and Mulford, W. (2002). Leadership and school Pearce, C.L. and Conger, J.A. (2003b). All those years ago: results. In Leithwood, K. and Hallinger, P. (eds), Second the historical underpinnings of shared leadership. In International Handbook of Educational Leadership and Pearce, C.L. and Conger, J.A. (eds), Shared Leadership: Administration. Dordrecht: Kluwer, pp. 561–612. Reframing the Hows and Whys of Leadership. Thousand Simkins, T. (2005). Leadership in education: ‘what works’or Oaks, CA: Sage Publications, pp. 1–18. ‘what makes sense’? Educational Management Adminis- Pearce, C.L., Conger, J.A. and Locke, E.A. (2008). Shared tration & Leadership, 33, pp. 9–26. leadership theory. Leadership Quarterly, 19, pp. 622– Sinclair, A. (1992). The tyranny of a team ideology. Organ- 628. ization Studies, 13, pp. 611–626. Perkins, N. (1993). Person-plus: a distributed view of think- Spillane, J.P. (2006). Distributed Leadership. San Francisco, ing and learning. In Salomon, G. (ed.), Distributed Cog- CA: Jossey-Bass. nitions: Psychological and Educational Considerations. Spillane, J. and Diamond, J.B. (2007a). Distributed Leader- Cambridge UK: Cambridge University Press, pp. 88–110. ship in Practice. New York: Teachers College Press. Pondy, L.R. (1978). Leadership as a language game. In Spillane, J.P. and Diamond, J.B. (2007b). Taking a distrib- McCall, M. and Lombardo, M. (eds), Leadership: Where uted perspective. In Spillane, J.P.and Diamond, J.B. (eds), Else Do We Go? Durham, NC: Duke University Press, pp. Distributed Leadership in Practice. New York: Teachers 87–99. College Press, pp. 1–15. Pye, A. and Knight, L. (2005). Network leadership: a new Spillane, J.P., Halverson, R. and Diamond, J.B. (2004). perspective on leadership? Paper presented at the Interna- Towards a theory of leadership practice: a distributed tional Conference on Studying Leadership, University of perspective. Journal of Curriculum Studies, 36, pp. 3– Lancaster, 12–13 December. 34. Ray, T., Clegg, S. and Gordon, R. (2004). A new look at Stoll, L. and Louis, K.S. (2007). Professional Learning dispersed leadership: power, knowledge and context. In Communities. New York: Open University Press. Storey, J. (ed.), Leadership in Organizations: Current Storey, A. (2004). The problem of distributed leadership in Issues and Key Trends. London: Routledge, pp. 319–336. schools. School Leadership & Management, 24, pp. 249– Resnick, L.B. (1991). Shared cognition: thinking as social 265. practice. In Resnick, L.B., Levine, J.M. and Teasley, S.D. Sugrue, C. (2009). From heroes and heroines to hermaph- (eds), Perspectives on Socially Shared Cognition. Wash- rodites: emasculation or emancipation of school leaders ington DC: American Psychological Association, pp. and leadership? School Leadership and Management, 29, 1–22. pp. 353–371. Rippin, A. (2007). Stitching up the leader: empirically based Sveningsson, S. and Alvesson, M. (2003). Managing mana- reﬂections on leadership and gender. Journal of Organ- gerial identities: organizational fragmentation, discourse izational Change Management, 20, pp. 209–226. and identity struggle. Human Relations, 56, pp. 1163– Robinson, V.M.J. (2009). Fit for purpose: an educationally 1193. relevant account of distributed leadership. In Harris, A. Timperley, H. (2005). Distributed leadership: developing (ed.), Distributed Leadership. Dordrecht: Springer, pp. theory from practice. Journal of Curriculum Studies, 37, 219–240. pp. 395–420. Rosenholtz, S.J. (1989). Teachers’ Workplace: The Social Torrance, D. (2009). Distributed leadership in Scottish Organization of Schools. New York: Longman. schools: perspectives from participants recently complet- Rosenthal, C.S. (1998). Determinants of collaborative leading the revised Scottish Qualiﬁcation for Headship Pro- ership: civic engagement, gender or organizational gramme. Management in Education, 23, pp. 63–70. norms? Political Research Quarterly, 51, pp. 847–868. Uhl-Bien, M. (2006). Relational leadership theory: explor- Ross, L., Rix, M. and Gold, J. (2005a). Learning distributed ing the social processes of leadership and organizing. leadership: part 1. Industrial and Commercial Training, Leadership Quarterly, 17, pp. 654–676. 37, pp. 130–137. Uhl-Bien, M., Marion, R. and McKelvey, B. (2007). Com- Ross, L., Rix, M. and Gold, J. (2005b). Learning distributed plexity leadership theory: shifting leadership from the leadership: part 2. Industrial and Commercial Training, industrial age to the knowledge era. Leadership Quar- 37, pp. 224–231. terly, 18, pp. 298–318.

Vygotsky, L.S. (1978). Mind in Society: The Development organizational landscape. Educational Management of Higher Psychological Processes. Cambridge, MA: Administration and Leadership, 37, pp. 430–451. Harvard University Press. Yammarino, F. and Dansereau, F. (2008). Multi-level nature Washbush, J.B. (2005). There is no such thing as leadership, of and multi-level approaches to leadership. Leadership revisited. Management Decision, 43, pp. 1078–1085. Quarterly, 19, pp. 135–141. Weick, K.E. and Roberts, K.H. (1993). Collective mind in Yeung, S.W.A., Lee, Y. and Yue, K.W.R. (2006). Multi- organizations: heedful interrelating on ﬂight decks. cultural leadership, sustainable total school environment. Administrative Science Quarterly, 38, pp. 357–381. Educational Research for Policy and Practice, 5, pp. 121– Wertsch, J.V. (1991). Voices of the Mind: A Sociocultural 131. Approach to Mediated Action. Cambridge, MA: Harvard York-Barr, J. and Duke, K. (2004). What do we know about University Press. teacher leadership? Findings from two decades of schol- Wheatley, M. (1994). Leadership and the New Science: arship. Review of Educational Research, 74, pp. 255–316. Discovering Order in a Chaotic World. San Francisco, Youngs, H. (2009). (Un)critical times? Situating distributed CA: Berrett-Koehler. leadership in the ﬁeld. Journal of Educational Adminis- Woods, P. (2004). Democratic leadership: drawing distinc- tration and History, 41, pp. 377–389. tions with distributed leadership. International Journal of Yukl, G.A. (2002). Leadership in Organizations, 5th edn. Leadership in Education, 7, pp. 3–26. Upper Saddle River, NJ: Prentice-Hall. Woods, P.A. and Gronn, P.(2009). Nurturing democracy: the contribution of distributed leadership to a democratic